1. Volume 13, Issue 5, Pages 601-611 (May 2011)
Mitochondrial Matrix Calcium Is an Activating Signal for Hormone Secretion 
Andreas Wiederkehr, Gergő Szanda, Dmitry Akhmedov, Chikage Mataki, Claus W. Heizmann, Kristina Schoonjans, Tullio Pozzan, András Spät, Claes B. Wollheim 
Cell Metabolism 
Volume 13, Issue 5, Pages (May 2011)
DOI: /j.cmet
Copyright © 2011 Elsevier Inc. Terms and Conditions

2. Figure 1 Expression and Targeting of S100G to the Mitochondrial Matrix
Insulinoma INS-1E (A–C) or adrenocortical H295R cells (D) were infected with adenoviruses for the expression of mitoS100G.
(A and B) Protein extracts were analyzed by western blotting. mitoS100G was induced with increasing concentrations of DOX for 24 hr (A) or for different times in the presence of 1 μg/ml DOX (B). The blots were probed with an anti-S100G (upper panels) and reprobed with an anti-GAPDH antibody (lower panels).
(C and D) Cells were stained with anti-S100G (red). H295R cells were loaded with MitoTracker dye for 30 min prior to fixation (D; green). Colocalization is shown in yellow. Scale bar = 5 μm. GAPDH, glyceraldehyde 3-phosphate dehydrogenase. See also Figures S1 and S2.
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions

3. Figure 2
Specific Buffering of Mitochondrial Matrix Ca2+ Rises with mitoS100G
(A–C) HeLa cells were perifused in KRBH buffer containing 5.5 mM glucose and stimulated with 1 μM histamine.
(D–F) INS-1E cells were perifused in KRBH buffer containing 2.5mM glucose. Ca2+ influx was stimulated by using 30 mM KCl.
(A–F) Cells were infected with adenoviruses for the expression of mitoS100G (A, B, and D–F) or mitoECFP (C) and either an adenovirus carrying mitochondrial aequorin (A, C, and D) or cytosolic aequorin (B and E). After infection, the cells were cultured for 24 hr in the presence of 1 μg/ml DOX (induction of mitoS100G or mitoECFP; gray traces) or absence of DOX (control; black traces). For experiments with HeLa cells the aequorin variants were expressed constitutively driven by the chicken actin promoter. The average ± SEM from n = 5 (A), n = 4 (B, C, and E), and n = 3 (D and F) experiments is shown. See also Figure S2.
(F) Static insulin secretion. Prior to the experiments, INS-1E cells were infected with Ad-mitoS100G in combination with Ad-tetON and cultured for 24 hr in the presence or absence of 1 μg/ml DOX. The cells were incubated for 30 min at 37°C in KRBH containing 2.5 mM glucose or stimulated with 30 mM KCl.
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions

4. Figure 3
Buffering [Ca2+]m Rises Lowers Glucose-Stimulated Insulin Secretion
(A–D) INS-1E cells were infected with Ad-mitoS100G in combination with Ad-tetON (A and B) or Ad-mitoS100A1 plus Ad-tetON (C and D) in combination with Ad-mitoAequorin for [Ca2+]m measurements (A and C). After infection, the cells were grown in INS medium in the absence (control cells, black traces) or presence of 1 μg/ml DOX (induced transgene, gray traces) for 24 hr. Cells were stimulated by raising glucose from 2.5 to 16.7 mM. For clarity, error bars are only shown every 120 s. (A–D) Average ± SEM; n = 3.
(B and D) Static insulin secretion experiments. Insulin secretion and content were measured in INS-1E cells expressing mitoS100G (B) or mitoS100A1 (D) after incubation for 30 min at 37°C in KRBH containing either 2.5 mM or 16.7 mM glucose. See also Figure S3.
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions

5. Figure 4
Mitochondrial Matrix Ca2+ Buffering Does Not Inhibit Cytosolic Ca2+ Responses after Glucose Stimulation
(A–E) INS-1E cells were infected with Ad-mitoS100G in combination with Ad-tetON and Ad-RIP-cytoAequorin (A) or Ad-RIP-YC3.6 (B–E) to measure [Ca2+]c. After infection, INS-1E cells were cultured for 24 hr with (mitoS100G; gray trace) or without (control; black trace) 1 μg/ml DOX. The average glucose-induced increase of [Ca2+]c from three independent experiments (± SEM) is shown.
(B and C) Ca2+ transients in response to glucose in a control (B) or mitoS100G expressing cell (C).
(D) The average peak frequency at 2.5 mM, 16.7 mM glucose, and in response to 200 μM tolbutamide (average ± SEM; n = 6). A total of 31 control infected cells and 43 mitoS100G-expressing cells were analyzed.
(E) MitoS100G (left, red) expression in YC3.6-positive cells (right, yellow) was confirmed by immunocytochemistry. Scale bar = 5 μm.
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions

6. Figure 5
Mitochondrial Matrix Ca2+ Signals Are Required for the Sustained Activation of Mitochondrial Energy Metabolism during Glucose Stimulation of INS-1E Cells
Glucose-mediated Ca2+ signaling in INS-1E cells was either suppressed by removing extracellular Ca2+ (KRBH without added Ca2+ plus 0.4 mM EGTA; A and D) or specifically buffered in the mitochondria by using mitoS100G (C, E, and F). Respiration rates (A–C) were measured every 9 min.
(A) Cells were stimulated with 16.7 mM glucose in the presence (filled diamonds) or absence of extracellular Ca2+ (empty squares). Because of variations in the absolute respiration rates, one typical experiment of three independent experiments performed in quadruplicate is shown (average ± SEM).
(B) Glucose-induced respiratory response of control uninfected INS-1E cells (filled diamonds) or 24 hr after infection with Ad-mitoS100G in combination with Ad-tetON (empty squares) in the absence of DOX.
(C) As described for (B) but in the presence of DOX (1 μg/ml) for 24 hr.
(B and C) The average of three independent experiments (± SEM), each performed in quadruplicate, is shown.
(D) Glucose-induced ATP changes in control (black trace) and after chelation of extracellular Ca2+ (gray trace).
(D and E) For cytosolic ATP measurements, INS-1E cells were coinfected with Ad-RIP-Luciferase. Representative examples of four independent experiments are shown.
(E) Glucose response 24 hr after infection with Ad-mitoS100G in combination with Ad-tetON cultured with (gray trace) or without (black trace) 1 μg/ml of DOX.
(F) Glucose-dependent hyperpolarization of the mitochondrial membrane potential measured with JC-1. Virus infection without induction of the transgene (control; upper black trace) or after 24 hr of mitoS100G expression (gray trace). Lower black trace: infected cells maintained at 2.5 mM glucose. FCCP (10 μM) was added to depolarize the mitochondrial membrane potential (mean ± SEM; n = 3).
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions

7. Figure 6
Matrix Ca2+ Buffering in Primary Rat Islet Cells Blunts Second-Phase Insulin Secretion
Substrate-attached primary rat islet cells were infected with Ad-mitoS100G in combination with Ad-tetON and Ad-mitoAequorin 24 hr before the experiment. Islet cells were perifused and [Ca2+]m (A and B) and insulin secretion (C and D) were measured on the same cell preparations. Islet cells were stimulated with 16.7 mM glucose. Depolarization was induced with 30 mM KCl.
(A) [Ca2+]m responses were measured in control islet cells (without induction of mitoS100G; black trace) or expressing mitoS100G (1 μg/ml DOX; 24 hr; gray trace). For clarity standard error bars are only shown every 120 s.
(B) Quantification of the [Ca2+]m response (area under the curve).
(C) Insulin concentration in fractions during islet cell perifusion described in (A).
(D) Quantification of insulin secretion. First phase: insulin secreted during 6 min after initiation of the glucose response. Second phase: sum 12–24 min after initiation of the glucose response. 30mM KCl: the sum of insulin secreted during 4 min of KCl-induced depolarization is shown.
(B and D) Black bars show control and gray bars show values after induction of mitoS100G (1 μg/ml of DOX). The average ± SEM of five independent experiments is shown.
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions

8. Figure 7
mitoS100G Lowers the Angiotensin II-Stimulated [Ca2+]m Response, NAD(P)H Formation, and Aldosterone Secretion
H295R cells were infected with Ad-mitoS100G in combination with Ad-tetON and loaded 48 hr later with Fluo-4 (cytosolic Ca2+) and Rhod-2 (mitochondrial Ca2+).
(A) Cytosolic (black trace) and mitochondrial (gray trace) Ca2+ rises in response to angiotensin II (1 nM) were measured simultaneously in individual cells. Maximal fluorescence changes were expressed as fold increase over basal values for both parameters.
(B) Analysis of average cytosolic (black bars) and mitochondrial (gray bars) Ca2+ responses to angiotensin II. Control infected cells (n = 10) were compared to cells expressing mitoS100G (n = 20) after 48 hr induction with 1 μg/ml of DOX.
(C) NAD(P)H autofluorescence increases in response to 1 nM angiotensin II were measured in control (33 cells) and mitoS100G-expressing cells (33 cells).
(D) Basal (black bars) and angiotensin II-stimulated hormone secretion (gray bars) was measured from control infected cells and cells after induction (1 μg/ml DOX) of mitoS100G for 48 hr (n = 6).
(A–D) The average ± SEM is shown.
Cell Metabolism  , DOI: ( /j.cmet )
Copyright © 2011 Elsevier Inc. Terms and Conditions